Infection with human immunodeficiency virus confers the greatest known risk of reactivation of a latent tuberculous infection and of progression of a recently acquired infection to disease. Therefore, in regions of high prevalence of HIV and M. tuberculosis, an increase in cases of tuberculosis has been and will be a major result of the AIDS pandemic. Uganda is among the countries with the highest prevalence in the world of infection with HIV . Over 1 million of the 17 million population are infected. Given the prevalence of infection with M. tuberculosis of 50-- 70%, approximately 50,000 excess cases of tuberculosis are predicted to occur each year because of the HIV epidemic. This unfortunate circumstance, nonetheless, provides the opportunity for a comprehensive investigation of the interactions of these infectious agents. Current understanding of HIV/M. tuberculosis interactions has focussed on the effects of HIV on reactivation of latent M. tuberculosis infection. There is, however, experimental and clinical evidence to support the hypothesis that tuberculosis by activating the cells harboring HIV- infection and by concomitant immunosuppression favors viral replication and feeds into a vicious cycle producing morbidity and shortening survival. The specific aim of this proposal is to perform a prospective epidemiologic study in Kampala, Uganda to test the hypothesis that active tuberculosis, even when successfully treated, promotes HIV activity leading to increased morbidity and shortened survival. In this study, we will compare the survival of HIV-infected patients with treated tuberculosis (cases) and HIV-infected individuals without active tuberculosis (controls) who are matched on age, gender, and CD4 lymphocyte count. In addition, we will obtain and preserve specimens for quantitative virologic assays and immunologic determinations in cases and controls. If the hypothesis is correct, then prevention of tuberculosis theoretically not only would reduce the risk of developing active tuberculosis in an HIV-infected individual but also prevent the activation of latent virus, thereby leading to prolonged survival. The results of the study, therefore, may provide further impetus for the application of preventive therapy for tuberculosis. The team of investigators has experience in conducting longitudinal studies in Uganda and is qualified to accomplish the research goals of this project. This is a health-related project.